Glyphosate for winter weed control in lucerne (Medicago sativa L.) : A dissertation submitted in partial fulfilment of the requirement for the Degree of Bachelor of Agricultural Science with Honours at Lincoln University

As lucerne stands age the plant population declines and perennial weed species such as dock, dandelion, twitch, white clover and ryegrass invade the resident lucerne. Existing winter weed control herbicides such as paraquat and atrazine are expensive and often inefficient at controlling these weeds. Therefore, three field experiments were conducted in Iversen Field, Lincoln University, to investigate the effectiveness of four rates (0, 1, 2 & 4 L/ha) of the herbicide glyphosate as a perennial winter weed control option for lucerne (Medicago sativa L.). The objectives of this research were to determine the optimum date and rate for the use of glyphosate in an 8 year old lucerne stand, by quantifying the phytotoxicity effects of glyphosate on lucerne and the target weed species. Experiment 1 was topped to remove herbage on the 4/4/2019, with the glyphosate applied on the 29/4/2019. The delay between herbage removal and herbicide application allowed the regrowth of plant herbage (kg DM/ha). Thus, the application of herbicide was onto actively growing plants in Experiment 1 which increased effects of phytotoxicity on the lucerne and reduced its yield. Target weed species were suppressed, particularly by the 2 L/ha treatment. However, the suppression of twitch, dock and clover was temporary however. Experiment 2 was topped on the 31/5/2019 and the glyphosate applied on the 6/6/2019. Minimal lucerne phytotoxicity occurred, due to its dormancy and the low proportion of green leaf present on the stand at the time of application. Sufficient weed suppression was achieved of all plant species, particularly from the 2 L/ha treatment which reduced the weed content from 51% to 17%. Experiment 3 was topped on the 23/7/2019 and the herbicide was applied on the 24/7/2019. Severe phytotoxicity effects were observed on all plant species. This late application delayed spring regrowth and gave insufficient feed supply. The objectives of this study were met by concluding the optimum rate and date for glyphosate application on lucerne was 2 L a.i./ha at the beginning of June. From this date and rate, minimal phytotoxicity effects were observed from the lucerne as no herbage was present at the time application, and the perennial weeds were suppressed. The addition of atrazine to the glyphosate to control opportunistic spring annuals requires further investigation.